INDEX

Explanations

viral

np_max-act · gemini-2.0-flash

The neuron activates on medical terms referring to infections (e.g. “viral,” “bacterial,” “infection”) and their treatments (e.g. “antibiotics,” “antiviral”), i.e. it detects mentions of infectious diseases and related therapies.

oai_token-act-pair · o4-mini Triggered by @xinyanhu8

New Auto-Interp

Configuration

andyrdt/saes-llama-3.1-8b-instruct/resid_post_layer_11/trainer_1

Dataset (Dashboard)

Various

Features

131,072

Data Type

float32

Hook Name

blocks.11.hook_resid_post

Architecture

standard

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 indexPath

-0.07

candidates

-0.07

>p

-0.06

 films

-0.06

 tangent

-0.06

 collapse

-0.06

 constraints

-0.06

Zero

-0.06

プ

-0.06

wcs

-0.06

POSITIVE LOGITS

里的

0.06

上了

0.06

елю

0.06

Included

0.06

 vnitř

0.06

سین

0.06

\Models

0.06

Implementation

0.06

toEqual

0.06

ves

0.06

Activations Density 0.009%

viral

The neuron activates on medical terms referring to infections (e.g. “viral,” “bacterial,” “infection”) and their treatments (e.g. “antibiotics,” “antiviral”), i.e. it detects mentions of infectious diseases and related therapies.

No Comments

No Known Activations

viral

The neuron activates on medical terms referring to infections (e.g. “viral,” “bacterial,” “infection”) and their treatments (e.g. “antibiotics,” “antiviral”), i.e. it detects mentions of infectious diseases and related therapies.

No Comments

No Known Activations